Wind and Structures

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

Reconstruction of wind speed fields in mountainous areas using a full convolutional neural network

  • Ruifang Shen (School of Civil Engineering, Chongqing University) ;
  • Bo Li (School of Civil Engineering, Chongqing University) ;
  • Ke Li (School of Civil Engineering, Chongqing University) ;
  • Bowen Yan (School of Civil Engineering, Chongqing University) ;
  • Yuanzhao Zhang (School of Civil Engineering, Beijing Jiaotong University)
  • Received : 2023.06.01
  • Accepted : 2024.01.23
  • Published : 2024.04.25
⟨ Previous Next ⟩

Abstract

As wind farms expand into low wind speed areas, an increasing number are being established in mountainous regions. To fully utilize wind energy resources, it is essential to understand the details of mountain flow fields. Reconstructing the wind speed field in complex terrain is crucial for planning, designing, operation of wind farms, which impacts the wind farm's profits throughout its life cycle. Currently, wind speed reconstruction is primarily achieved through physical and machine learning methods. However, physical methods often require significant computational costs. Therefore, we propose a Full Convolutional Neural Network (FCNN)-based reconstruction method for mountain wind velocity fields to evaluate wind resources more accurately and efficiently. This method establishes the mapping relation between terrain, wind angle, height, and corresponding velocity fields of three velocity components within a specific terrain range. Guided by this mapping relation, wind velocity fields of three components at different terrains, wind angles, and heights can be generated. The effectiveness of this method was demonstrated by reconstructing the wind speed field of complex terrain in Beijing.

Keywords

Acknowledgement

This paper is supported by China national key research and development plan (2019YFF0301904), Science Fund for Creative Research Groups of the National Natural Science Foundation of China(52221002), the 111 project of the Ministry of Education and the Bureau of Foreign Experts of China (B13002, B18062), Key Laboratory of Wind Resistance Technology of Bridge Structure and Transportation Industry (Tongji University) open project (KLWRTBMC22-01), the Natural Science Foundation of Chongqing, China (cstc2020jcyj-msxmX0773), and the Fundamental Research Funds for the Central Universities (2020CDJ-LHZZ-018).

References

  1. Breckling, J. (2012), "The analysis of directional time series: applications to wind speed and direction", Springer Science and Business Media, 61.
  2. Castro, F.A., Palma, J.M.L.M. and Lopes, A.S. (2003), "Simulation of the Askervein flow. Part 1: Reynolds averaged Navier-Stokes equations (k-ɛ turbulence model)", Bound. Lay. Meteorol., 107(3), 501-530. https://doi.org/10.1023/A:1022818327584.
  3. Getter, D., Bessac, J., Rudi, J. and Feng, Y. (2023), "Statistical treatment of convolutional neural network super-resolution of inland surface wind for subgrid-scale variability quantification", arXiv:2211.16708v3. https://doi.org/10.48550/arXiv.2211.6708.
  4. Guo, X., Li, W. and Iorio, F. (2016), "Convolutional neural networks for steady flow approximation", In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'16), ACM, New York, Y, USA, 2016.
  5. Hesthaven, J.S. and Ubbiali, S. (2018), "Non-intrusive reduced order modeling of nonlinear problems using neural networks", J. Comput. Phys., 363, 55-78. https://doi.org/10.1016/j.jcp.2018.02.037
  6. Hoolohan, V., Tomlin, A.S. and Cockerill, T. (2018), "Improved near surface wind speed predictions using Gaussian process regression combined with numerical weather predictions and observed meteorological data", Renew. Energy, 126(10), 1043-1054. https://doi.org/10.1016/j.renene.2018.04.019.
  7. Jin, X.W., Cai, S.Z., Li, H. and Karniadakis, G.E. (2021), "NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible NavierStokes equations", J. Comput. Phys., 426, 109951.
  8. LeCun, Y., Bengio, Y. and Hinton, G. (2015), "Deep learning", Nature, 521(7553), 436-444. https://doi.org/10.1038/nature14539.
  9. LeCun, Y., Boser, B., Denker, J.S., Henderson, D., Howard, R.E., Hubbard, W. and Jackel, L.D. (1989), "Backpropagation applied to handwritten zip code recognition", Neural Comput., 1(4), 541-551. https://doi.org/10.1162/neco.1989.1.4.541.
  10. Long, J., Shelhamer, E. and Darrell, T. (2015), "Fully convolutional networks for semantic segmentation", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 3431-3440. https://doi.org/10.1109/CVPR.2015.7298965.
  11. Ma, W.J, Zhang, J. and Yu J. (2021), "Non-intrusive reduced order modeling for flowfield reconstruction based on residual neural network", Acta Astronautica, 183, 346-362. https://doi.org/10.1016/j.actaastro.2020.11. 050.
  12. Maurizi, A., Palma, J.M.L.M. and Castro, F.A. (1998), "Numerical simulation of the atmospheric flow in a mountainous region of the North of Portugal", J. Wind Eng. Ind. Aerod., 74(6), 219-228. https://doi.org/10. 1016/S0167-6105(98)00019-1. https://doi.org/10.1016/S0167-6105(98)00019-1
  13. Palma, J.M.L.M., Castro, F.A., Ribeiro, L.F., Rodrigues, A.H. and Pinto, A.P. (2008), "Linear and nonlinear models in wind resource assessment and wind turbine micro-siting in complex terrain", J. Wind Eng. Ind. Aerod., 96(12), 2308-2326. https://doi.org/10.1016/j. jweia. 2008.03.012.
  14. Qin, L., Liu, S., Long, T., Shahzad, M.A., Schlabergm H.I. and Yan, S.A. (2018), "Wind field reconstruction using dimension-reduction of CFD data with experimental validation", Energy, 151, 272-288. https://doi.org/10.1016/j.energy.2018.02.141.
  15. Raissi, M., Perdikaris, P. and Karniadakis, G.E. (2019), "Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations", J. Comput. Phys., 378, 686-707. https://doi.org/10.1016/j.jcp.2018.10.045
  16. Saavedra-Moreno, B., De La Iglesia, A., Magdalena-Saiz, J., Carro-Calvo, L., Duran, L. and Salcedo-Sanz, S. (2015), "Surface wind speed reconstruction from synoptic pressure fields: machine learning versus weather regimes classification techniques", Wing Energy, 18(9), 1531-1544. https://doi.org/10.1002/we.1774.
  17. Sun, S., Liu, S., Liu, J. and Schlaberg, H.I. (2019), "Wind field reconstruction using inverse process with optimal sensor placement", IEEE Trans. Sustain. Energy, 10(3), 1290-1299. https://doi.org/10.1109/TSTE.2018.2865512
  18. Uchida, T. and Ohya, Y. (2008), "Micro-siting technique for wind turbine generators by using large-eddy simulation", J. Wind Eng. Ind. Aerod., 96(10-11), 2121-2138. https://doi.org/10.1016/j.jweia.2008.02.047.
  19. Xue, Y.S., Chen, N., Wang, S.M., Wen, F.H., Lin, Z.Z. and Wang, Z. (2017), "Review on wind speed prediction based on spatial correlation", Autom. Electric Power Syst., 41(10), 161-169. (in Chinese) https://doi.org/10.19912/j.0254-0096.tynxb.2020-0410.
  20. Yan, B.W., Shen, R.F., Li, K., Wang, Z.G., Yang, Q.S., Zhou, X.H. and Zhang, L. (2023), "Spatio-temporal correlation for simultaneous ultra-short-term wind speed prediction at multiple locations", Energy, 284, 128418. https://doi.org/10.1016/j.energy.2023.128418.
  21. Zeiler, M.D. and Fergus, R. (2014), "Visualizing and Understanding Convolutional Networks", ECCV 2014: Part I/Springer,, 2014:818-833.
  22. Zeiler, M.D., Taylor, G.W. and Fergus, R. (2011), "Adaptive deconvolutional networks for mid and high level feature learning", International Conference on Computer Vision, 2018-2025. https://doi.org/10.1109/ICCV.2011.6126474.
  23. Zhang, G.C., Zheng, X.X., Liu, S., Chen, M.X., Wang, C.Q. and Wang, X.Y. (2022), "Three-dimensional wind velocity reconstruction based on tensor decomposition and CFD data with experimental verification", Energy Conver. Manage., 256, 115322. https://doi.org/10.1016/j.enconman.2022.115322.